LambdaHack VS rattletrap

Looks really cool! Here's how I do releases: https://github.com/LambdaHack/LambdaHack/issues/76

here's the failed commit: https://github.com/LambdaHack/LambdaHack/commit/930dfd46b949d1525e900b997137389a8092305b

I've recently tried that and gave up (https://github.com/LambdaHack/LambdaHack/issues/248). Perhaps it's possible to statically link SDL2, but I learnt it's impossible to statically link the OpenGL and X11 libraries, so you end up with a partially statically linked binary (I didn't manage to obtain even that).

I did some plotline sketches on the github wiki of my project, but never used them so far. Not much use from the other musings there, either. Instead I tweak the templates for procedural generation directly in the code (game content part of the code, to be precise) and get a lot of mileage out of that. The only page of the wiki I got lots of benefit from is the derivation of the simplified formula for calculating speed, distance and damage of projectile from their weight: https://github.com/LambdaHack/LambdaHack/wiki/Item-statistics#projectile-velocity-and-distance

Around a decade. No, one release per 1-2 years. The differences are quite large, both gameplay and, even more, UI, in particular in using ready fornts, then custom fonts, then many fonts at once. See the changelogs (that's only for the egine, but that's where most of the changes are made): https://github.com/LambdaHack/LambdaHack/releases

rattletrap works quite well and is still actively maintained, the lack of stats is because we're a relatively small e-sport and stats are hard, meaning there just isn't that many people working on it - but not because of the tooling

This is the one (rattletrap) that can convert it back to replay as well (harder to read): https://github.com/tfausak/rattletrap/releases/tag/11.2.4

Ah I see that's true. There is this (GitHub)[https://github.com/tfausak/rattletrap] that lets you edit stuff in replays. I haven't really fiddled too much with it, I really don't have a reason to edit replays. My speculation on the programming is that it uses AI techniques to assume what action you are going to do next based on your prior movements and tries to reconcile it in the event of latency mitigation or an error on their end.

But you need to first convert the replay to json, and then you can start to replace stuff. Then convert it back and it works. There are softwares that does the converting. When I was experimenting with those I tried 3 different replay parsers and I liked this the most: https://github.com/tfausak/rattletrap